During an earthquake, fault surfaces slip at speeds of several cm/s to several m/s. The frictional properties at these slip rates play an important role in determining key aspects of seismic rupture, including earthquake stress drop, heat production, and the mode of dynamic rupture expansion. However, little is known about friction at dynamic slip velocities. The limited available data have been collected under a narrow range of conditions and significant discrepancies exist between these data and the friction database at lower velocities appropriate to earthquake nucleation. This proposal is to conduct laboratory experiments spanning slip speeds from mm/s to m/s. An important goal will be to quantitatively connect friction observations over the entire velocity range and to study the problem of scaling laboratory friction data to seismogenic faults. Bare rock surfaces and granular fault gouge will be studied under a range of normal stresses and surface conditions, and friction data will be analyzed in terms of existing friction theory. A major goal of the experimental program will be to identify the micromechanical causes of velocity-dependent friction and friction evolution (state) effects.
